PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Share a file Manage my documents Convert Recover PDF Search Help Contact



Am. J. Epidemiol. 2014 de Jong 1323 30 .pdf



Original filename: Am. J. Epidemiol.-2014-de Jong-1323-30.pdf
Title: untitled

This PDF 1.5 document has been generated by Arbortext Advanced Print Publisher 10.0.1465/W Unicode / Acrobat Distiller 7.0 (Windows), and has been sent on pdf-archive.com on 15/08/2014 at 20:02, from IP address 189.120.x.x. The current document download page has been viewed 475 times.
File size: 161 KB (8 pages).
Privacy: public file




Download original PDF file









Document preview


American Journal of Epidemiology
© The Author 2014. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of
Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Vol. 179, No. 11
DOI: 10.1093/aje/kwu053
Advance Access publication:
April 29, 2014

Original Contribution
Association of Occupational Pesticide Exposure With Accelerated Longitudinal
Decline in Lung Function

Kim de Jong, H. Marike Boezen, Hans Kromhout, Roel Vermeulen, Dirkje S. Postma, and
Judith M. Vonk*

Initially submitted November 19, 2013; accepted for publication February 25, 2014.

Cross-sectional studies have shown that occupational exposure to vapors, gases, dusts, and fumes (VGDF) and
pesticides is associated with a lower level of lung function. These associations seem to be stronger in ever smokers.
In the current study, we aimed to assess whether occupational exposure to VGDF and pesticides is associated
with longitudinal decline in lung function. We used 12,772 observations from 2,527 participants in the VlagtweddeVlaardingen Study, a general-population-based cohort study that followed subjects for 25 years, from 1965 to the
last survey in 1989/1990. Job-specific exposure was estimated with the ALOHA+ job exposure matrix. Associations
between exposures and annual changes in forced expiratory volume in 1 second (FEV1) and FEV1 as a percentage
of inspiratory vital capacity (FEV1%VC) were assessed with linear mixed-effect models including sex, age, and level
of lung function at the first measurement and pack-years of smoking at the last measurement. We tested for interaction between smoking and occupational exposure and assessed associations separately for never smokers and
ever smokers. Exposure to VGDF was not associated with accelerated lung function decline after adjustment for coexposure to pesticides. Exposure to pesticides, both in the last-held job and as a cumulative measure, was associated with accelerated decline in FEV1 and FEV1%VC, especially among ever smokers, where we found an excess
change in FEV1 of –6.9 mL/year (95% confidence interval: –10.2, –3.7) associated with high pesticide exposure.
general population; longitudinal studies; lung; lung function decline; occupational exposure; pesticides; smoking

Abbreviations: CI, confidence interval; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 second;
FEV1%VC, FEV1 as a percentage of inspiratory vital capacity; JEM, job exposure matrix; VGDF, vapors, gases, dusts, and fumes.

smokers (5, 6). Moreover, we recently showed that occupational exposure to pesticides was cross-sectionally associated
with a lower level of lung function and increased prevalence
of COPD in 2 general populations (5).
Several studies have focused on specific exposures within
single industries or companies, such as exposure to dusts and
gases in tunnel construction workers, exposure to organic
dust in pig farmers, and exposure to cotton dust versus silk
dust in textile workers (7–9). To date, only 1 study has assessed the association between occupational exposures, as estimated by means of the ALOHA job exposure matrix (JEM),
and longitudinal decline in lung function in a general population. In that study, using meta-analyzed data from 27

A lower level of lung function and accelerated decline in
lung function are associated with respiratory diseases such
as chronic obstructive pulmonary disease (COPD). Although
cigarette smoking is regarded as the most important environmental risk factor for impaired lung function, accelerated
lung function decline, and development of COPD, there are
additional environmental risk factors. Approximately 15%–
20% of all COPD cases have been attributed to occupational
exposures (1). Cross-sectional studies have shown that occupational exposure to vapors, gases, dusts, and fumes (VGDF)
is a risk factor for respiratory symptoms, a lower level of lung
function, and COPD in the general population (2–4). These
associations are suggested to be even stronger in ever
1323

Am J Epidemiol. 2014;179(11):1323–1330

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

* Correspondence to Dr. Judith M. Vonk, Department of Epidemiology, Faculty of Medical Sciences, University of Groningen,
University Medical Center Groningen, Hanzeplein 1, 9700 RB Groningen, the Netherlands (e-mail: j.m.vonk@umcg.nl).

1324 de Jong et al.

METHODS
Study population and measurements

The study population consisted of subjects who participated in the last survey (1989/1990) of the VlagtweddeVlaardingen Study, a prospective general-population-based
cohort study on the epidemiology of pulmonary diseases.
The cohort study, including subjects from a rural area in
the northeastern part of the Netherlands (Vlagtwedde) and
subjects from an urban area in the southwestern part of the
Netherlands (Vlaardingen), started in 1965. Participants were
followed for 25 years, with surveys performed every 3 years
(11, 12). In Vlaardingen, only participants who were included at baseline (1965 or 1969) were approached for
follow-up, whereas in Vlagtwedde new subjects aged 20–
65 years were invited to participate at every survey. The study
protocol was approved by the local university medical hospital ethics committee (University Medical Center Groningen,
Groningen, the Netherlands). All participants gave their written informed consent. In 1984, the Committee on Human
Subjects in Research of the University of Groningen reviewed the study and affirmed the safety of the protocol
and study design.
During each survey, information was collected by questionnaires, and spirometry was performed with a water-sealed
spirometer using a slow inspiration maneuver, according to
European Respiratory Society criteria. (For more information
about the inspiratory vital capacity measurements, see Web
Appendix 1, available at http://aje.oxfordjournals.org/.) Job
titles and descriptions reported at the time of the last survey
were used to estimate job-specific VGDF and pesticide exposures in the current job (or the last-held job, in case of current
unemployment), and exposure was categorized as none, low,
or high (scored 0, 1, or 2, respectively) using the ALOHA+
JEM (4, 5). Cumulative exposure was calculated as the number of intensity-years in 3 jobs: the last-held job, the previous
job, and the most important job before the previous job, multiplied by the intensity of exposure (low = 1 and high =
4). All job titles and descriptions were obtained from the

questionnaire administered at the last survey in 1989/1990
(the specific survey question can be found in Web Appendix
2). Information on numbers of years worked in the reported
jobs was available for 93% of the subjects.
Statistical analysis

Associations between the exposures and annual changes in
forced expiratory volume in 1 second (FEV1) and FEV1 as a
percentage of inspiratory vital capacity (FEV1%VC) were assessed with linear mixed-effect models. For each subject, the
linear mixed-effect model takes into account every available
survey. Only surveys performed at age 30 years or older were
included, because an individual’s maximal level of lung
function is assumed to have been reached before that age
and thereafter lung function is considered to be in the decline
phase (13). The linear mixed-effect models included sex, age
at the first measurement (centered at age 30 years), level of
lung function at the first measurement (absolute value centered at the population mean level), pack-years of smoking
at the last measurement, and their interaction with time
since first measurement. Time was defined as years since
first measurement. A random factor was assigned to the intercept and time. Statistical analyses were performed with Spotfire S-PLUS, version 8.1 (TIBCO Software Inc., Palo Alto,
California) and SPSS, version 20 (IBM Corporation, Armonk,
New York). P values less than 0.05 were considered statistically significant (tested 2-sided).
Because of substantial co-exposure between the specific
occupational exposures, the analysis of VGDF exposure additionally included adjustment for exposure to pesticides, and
conversely the analysis of pesticide exposure included adjustment for exposure to VGDF. A high level of exposure to pesticides was always accompanied by a high level of exposure
to VGDF in our sample; therefore, testing for interaction
between the 2 exposures was not possible. In an attempt to
disentangle the associations with exposure to VGDF and
pesticides, we created groups based on joint exposure to
VGDF and pesticides (unexposed, high exposure to VGDF
only, or high exposure to both VGDF and pesticides) and
assessed associations with annual change in FEV1 and
FEV1%VC.
Additionally, we assessed whether the associations between occupational exposures and change in lung function
were different for never and ever smokers and tested for
interaction.
RESULTS
Population characteristics

Population characteristics are shown in Table 1. A total
of 12,772 observations from 2,527 subjects were available
(the median number of observations per subject was 5; range,
1–8). Of all subjects, 53% were male, and the median age at
the last visit was 53 years. There were 2 times more ever
smokers than never smokers. One-third of the subjects were
occupationally exposed to high levels of VGDF (33%),
whereas exposure to high levels of pesticides was less common (12%). The median number of intensity-years, estimated
Am J Epidemiol. 2014;179(11):1323–1330

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

individual European centers, Sunyer et al. (10) did not find an
association between exposure to dusts, gases, and fumes or
the composite measure VGDF and decline in lung function
between 2 surveys administered on average 9 years apart.
This null finding may have been due to the relatively young
age of the study population (25–45 years) and the relatively
short follow-up time; hence, cumulative exposure may have
been too limited (10). With the current study, we extended the
findings of previous analyses by using a general population
cohort that has been followed for 25 years, with surveys
performed every 3 years. Moreover, we now have additional
estimates of occupational exposure to pesticides from the recently extended ALOHA JEM (ALOHA+). Therefore, the
aim of the current study was to assess whether occupational
exposures to VGDF and pesticides (comprising herbicides
and insecticides) were associated with longitudinal decline
in lung function in a general population, and secondly
whether these associations differed for never and ever
smokers.

Pesticide Exposure and Lung Function Decline 1325

Table 1. Characteristics of Subjects Who Participated in the Last Survey of the Vlagtwedde-Vlaardingen Study, the
Netherlands, 1989/1990
No.

No. of subjects

%

Median

Range

Mean (SD)

2,527

No. of observations

12,772

No. of visits per subject

5

1–8

16

0–25

At first measurement

35

30–70

At last measurement

53

30–80

Duration of follow-up, years
Male sex

1,343

53

Age, years

Smoking status
Ever smoker

877

35

1,650

65

Pack-years of smoking in ever smokers

20

1–262

Lung function level at first measurement
FEV1 % predicteda

91 (13)

FEV1%VC

77 (8)

Exposure to vapors, gases, dusts, and fumes
High exposure in the last-held job
b

Cumulative exposure >0 intensity-years

837

33

1,626

69

Cumulative exposure,c intensity-years

48

1–260

56

1–228

Exposure to pesticidesd
High exposure in the last-held job

298

12

Cumulative exposure >0 intensity-years

579

25

Cumulative exposure,c intensity-years

Abbreviations: FEV1, forced expiratory volume in 1 second; FEV1%VC, FEV1 as a percentage of inspiratory vital
capacity; SD, standard deviation.
a
Based on reference equations by Quanjer et al. (24).
b
Intensity-years were estimated as years of exposure weighted by intensity of exposure (low = 1, high = 4).
c
Among the exposed subjects (>0 intensity-years).
d
Both herbicides and insecticides.

as years of exposure weighted by intensity of exposure (low = 1,
high = 4), within the exposed subjects (>0 exposure years)
was 48 years (range, 1–260) for VGDF and 56 years (range,
1–228) for pesticides. The mean estimated change in lung
function in the whole sample was −18.0 mL/year (95% confidence interval (CI): −19.6, −16.5) for FEV1 and −0.08%/year
(95% CI: −0.11, −0.05) for FEV1%VC.

associated with excess change in FEV1 (Table 2). There
was a marginally significant association with cumulative exposure (intensity-years) to VGDF (Table 2). There was no
significant interaction between smoking and VGDF exposure
(Table 3).

VGDF exposure and annual change in lung function

The participants who were unexposed to pesticides had an
average annual change of −17.6 mL/year (95% CI: −19.1,
−16.0) in FEV1 and an average annual change of −0.07%/
year (95% CI: −0.10, −0.04) in FEV1%VC. Compared
with no exposure, occupational exposure to high levels of
pesticides in the last-held job was associated with an excess
change of −6.2 mL/year (95% CI: −8.6, −3.8) in FEV1
and a change of −0.09%/year (95% CI: −0.14, −0.04) in
FEV1%VC. This association remained present after adjustment for co-exposure to VGDF (Table 2). The negative association between occupational exposure to pesticides and
annual change in lung function was confirmed when we

The group that was unexposed to VGDF had an average
annual change of −17.2 mL/year (95% CI: −19.0, −15.4)
in FEV1 and −0.07%/year (95% CI: −0.10, −0.03) in
FEV1%VC. Compared with no exposure, high occupational
exposure to VGDF in the last-held job was significantly associated with an excess change in FEV1 (−4.0 mL/year, 95%
CI: −6.1, −2.0) but was not significantly associated with
change in FEV1%VC (−0.04%/year, 95% CI: −0.08, 0.00).
When adjusted for pesticide exposure, occupational exposure
to VGDF in the last-held job was no longer significantly
Am J Epidemiol. 2014;179(11):1323–1330

Pesticide exposure and annual change in lung function

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

Never smoker

1326 de Jong et al.
Table 2. Associations Between Occupational Exposure to Pesticides and Annual Change in FEV1 and FEV1%VC (Linear Mixed-Effect Modela),
Vlagtwedde-Vlaardingen Study, the Netherlands, 1965–1990a
FEV1%VC

FEV1
Exposure and Measure

No. of
Subjects

% of
Subjects

No. of
Observations

Excess
Change,
mL/year

95% CI

P Value

Excess
Change,
%/year

95% CI

P Value

VGDF
Exposure category
Noneb

958

38

4,643

0

Low

732

29

3,698

−0.6

−2.6, 1.5

0.59

−0.01

−0.05, 0.03

0.58

High

837

33

4,431

−1.8

−4.4, 0.7

0.15

0.00

−0.06, 0.05

0.88

2,359

69

11,895

−0.2

−0.4, 0.0

0.047

0.00

−0.01, 0.00

0.24

2,067

82

10,258

0

Cumulative exposure,
intensity-yearsc,d

Reference

0

Reference

Exposure category
Noneb

Reference

0

Reference

Low

162

6

855

−1.4

−4.9, 2.1

0.44

−0.01

−0.08, 0.06

0.73

High

298

12

1,659

−5.1

−8.0, −2.1

<0.001

−0.09

−0.15, −0.03

0.004

2,359

25

11,895

−0.3

−0.5, −0.1

0.007

−0.01

−0.01, 0.00

0.046

Cumulative exposure,
intensity-yearsc,d

Abbreviations: CI, confidence interval; FEV1, forced expiratory volume in 1 second; FEV1%VC, FEV1 as a percentage of inspiratory vital capacity;
VGDF, vapors, gases, dusts, and fumes.
a
The model adjusted for sex, age, and level of lung function at the first measurement, pack-years of smoking at the last measurement, and
co-exposure to VGDF or pesticides.
b
No exposure in the last-held job at the time of the last survey (1989/1990).
c
Excess annual change in lung function per 10 intensity-years; intensity-years were estimated as years of exposure weighted by intensity of
exposure (low = 1, high = 4).
d
Number of subjects exposed for >0 intensity-years as a percentage of the total number of subjects.
e
Both herbicides and insecticides.

used an estimate of cumulative exposure to pesticides
(intensity-years) (Table 2).
The annual changes in both FEV1 and FEV1%VC were
significantly larger in ever smokers with high pesticide exposure than in never smokers with high pesticide exposure;
the P values for interaction between smoking and high exposure to pesticides in the last-held job were 0.02 and 0.01 for
FEV1 and FEV1%VC, respectively, after adjustment for coexposure to VGDF and the VGDF-by-smoking interaction.
When the associations were assessed for never and ever
smokers separately, the associations between occupational
exposure to pesticides, both in the last-held job and as a cumulative measure (intensity-years), and change in FEV1 and
FEV1%VC remained present in ever smokers only (Table 3).
Joint exposure to VGDF and pesticides

High exposure to pesticides was always accompanied by
high exposure to VGDF in our sample; that is, there was no
“high exposure to pesticide only” group. The group with high
exposure to both VGDF and pesticides in the last-held
job had a significant excess change of −6.7 mL/year (95%
CI: −9.2, −4.1) in FEV1 and a change of −0.09%/year (95%
CI: −0.14, −0.04) in FEV1%VC, compared with the unexposed
group. There was also a significant difference between the
group with high exposure to both VGDF and pesticides and

the group with high exposure to VGDF only (change in
FEV1 = −4.7 mL/year (95% CI: −7.5, −1.9); change in
FEV1%VC = −0.08%/year (95% CI: −0.14, −0.03)) (Figure 1).
There was no difference in lung function change between the
group with only high exposure to VGDF and the unexposed
group (FEV1: −2.0 mL/year (95% CI: −4.1, 0.1); FEV1%
VC: −0.00%/year (95% CI: −0.05, 0.05)).
DISCUSSION

Our current study shows that occupational exposure to pesticides is associated with accelerated annual decline in FEV1
and FEV1%VC in this sample from the Dutch general population. To our knowledge, no other study to date has investigated
the association between occupational exposure to pesticides
and decline in lung function in a general population. Crosssectional studies have shown associations of specific types of
pesticides with the presence of chronic bronchitis in US farmers (14) and their spouses (15) and with lower levels of FEV1
and FVC in occupationally exposed farmers from Sri Lanka
(16) and South Korea (17). Recently, we have shown that occupational exposure to pesticides, as assessed with the ALOHA+
JEM, was associated with lower levels of FEV1 and FEV1%VC
and an increased prevalence of COPD in a cross-sectional
analysis of 2 Dutch general populations, the LifeLines cohort
and the current Vlagtwedde-Vlaardingen cohort (5). With the
Am J Epidemiol. 2014;179(11):1323–1330

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

Pesticidese

FEV1

FEV1%VC

Never Smokers
Exposure and Measure

Excess
Change,
mL/year

95% CI

Ever Smokers
P Value

Excess
Change,
mL/year

95% CI

Never Smokers
P Value

Excess
Change,
%/year

95% CI

Ever Smokers
P Value

Excess
Change,
%/year

95% CI

P Value

VGDF
Exposure category
Noneb

0

Reference

0

Reference

Low

−1.2

−4.2, 2.0

0.47

−0.1

−2.7, 2.6

0.96

−0.03

−0.10, 0.03

0.31

0.01

−0.05, 0.06

0.77

High

−3.4

−8.8, 2.0

0.22

−1.9

−4.8, 1.1

0.21

−0.06

−0.17, 0.05

0.29

0.00

−0.06, 0.06

0.90

−0.3

−0.6, 0.1

0.19

−0.2

−0.4, 0.0

0.08

0.00

−0.01, 0.01

0.83

0.00

−0.01, 0.00

0.12

Noneb

0

Reference

0

Reference

0

Reference

Low

2.1

−4.6, 8.8

0.54

−3.2

−7.6, 1.2

0.16

0.07

−0.06, 0.21

0.28

−0.05

1.9

−4.8, 8.7

0.58

−6.9

0.0

−0.6, 0.6

0.92

−0.4

Cumulative exposure,
intensity-yearsc,d

0

Reference

0

Reference

Pesticidese
Exposure category

−10.2, −3.7

<0.001

0.08

−0.05, 0.22

0.23

−0.13

−0.6, −0.1

0.005

0.00

−0.01, 0.02

0.64

−0.01

Reference
−0.14, 0.04
f

−0.20, −0.07
−0.01, 0.00

0.24
<0.001
0.03

Abbreviations: CI, confidence interval; FEV1, forced expiratory volume in 1 second; FEV1%VC, FEV1 as a percentage of inspiratory vital capacity; VGDF, vapors, gases, dusts, and fumes.
a
The model adjusted for sex, age, and level of lung function at the first measurement, pack-years of smoking at the last measurement, and co-exposure to VGDF or pesticides and included
the interaction between ever smoking and occupational exposure.
b
No exposure in the last-held job at the time of the last survey (1989/1990).
c
Excess annual change in lung function per 10 intensity-years; intensity-years were estimated as years of exposure weighted by intensity of exposure (low = 1, high = 4).
d
Number of subjects exposed for >0 intensity-years as a percentage of the total number of subjects.
e
Both herbicides and insecticides.
f
P < 0.05 for interaction between smoking status and occupational exposure.

Pesticide Exposure and Lung Function Decline 1327

High
Cumulative exposure,
intensity-yearsc,d

f

0

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

Am J Epidemiol. 2014;179(11):1323–1330

Table 3. Associations Between Occupational Exposure to Pesticides and Annual Change in FEV1 and FEV1%VC Among Never and Ever Smokers (Linear Mixed-Effect Modela),
Vlagtwedde-Vlaardingen Study, the Netherlands, 1965–1990

1328 de Jong et al.

A)

3,300
3,200
3,100

FEV1, mL

3,000
2,900
2,800
2,700
2,600

VGDF –, Pesticides –

2,500

VGDF +, Pesticides –

2,400

VGDF +, Pesticides +

0

5

10

15
20
Time, years

25

30

25

30

B)
78
77

FEV1% VC

76
75
74
73
VGDF –, Pesticides –

72

VGDF +, Pesticides –

71

VGDF +, Pesticides +

70
0

5

10

15
20
Time, years

Figure 1. Forced expiratory volume in 1 second (FEV1) (A) and
FEV1 as a percentage of inspiratory vital capacity (FEV1%VC)
(B) among persons with high occupational exposure to vapors,
gases, dusts, and fumes (VGDF) only (VGDF +, Pesticides −) and persons with joint high occupational exposure to VGDF and pesticides
(VGDF +, Pesticides +) as compared with unexposed persons
(VGDF −, Pesticides −), by time (years) since first survey after age
30 years, Vlagtwedde-Vlaardingen Study, the Netherlands, 1965–
1990. Linear mixed-effect models were used and included adjustment
for sex, age, and level of lung function at the first measurement and
pack-years of smoking at the last measurement. A significantly accelerated decline was found for the group with joint high exposure to
VGDF and pesticides compared with the unexposed group (P <
0.001 for both FEV1 and FEV1%VC) and for the joint-high-exposure
group compared with the VGDF-only group (P = 0.001 for FEV1; P =
0.003 for FEV1%VC).

current study, we extended these findings by showing that high
exposure to pesticides had clinically relevant associations, especially in ever smokers, where we found an excess change in
FEV1 of –6.9 mL/year compared with no exposure.
Subjects who were highly exposed to pesticides in our
sample included field crop and vegetable growers (72%),
mixed crop and animal producers (12%), gardeners, horticultural and nursery growers (15%), and tree and shrub crop

Am J Epidemiol. 2014;179(11):1323–1330

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

2,300

growers (1%). To exclude the possibility that the associations
were driven by crop-farming-related exposures that we did
not account for in the current study, we performed an additional analysis in which we excluded the agricultural workers
(i.e., the field crop and vegetable growers and the mixed crop
and animal producers). Associations between high exposure
to pesticides and level of change in FEV1 (−13.1 mL/year,
95% CI: −19.1, −7.1) and FEV1%VC (−0.19%/year, 95%
CI: −0.31, −0.06) were even stronger in the remaining
group, of which the majority were gardeners and horticultural
and nursery growers. In an additional analysis, we assessed
associations with 2 pesticide subcategories: insecticides and
herbicides (Web Tables 1–3). Associations with insecticides
were similar to those for all pesticides. The strongest association with FEV1 was seen for low exposure to herbicides, an
association that may have been driven by the gardeners and
horticultural and nursery growers.
Within our sample, almost all subjects with high pesticide
exposure (about 99%) originated from the rural area around
Vlagtwedde in the northeastern part of the Netherlands.
Approximately 80%–90% of the agricultural land around
Vlagtwedde was used for cultivation of crops during the
study period (1965–1990). During the 1960s and 1970s,
the majority of cultivated crops were cereals (about 50%),
and during the 1980s and 1990s they were potatoes (about
50%). Within the potatoes sector, dinitrophenol herbicides
were used until the 1980s, whereas the quaternary ammonium
herbicides (diquat and paraquat) became the most commonly
used herbicides from the early 1980s onward (M. Brouwer,
University of Utrecht, personal communication, 2014; for
more detailed information, see Web Table 4).
Exposure to pesticides in the occupational setting occurs
during mixing, loading of equipment, and spraying and application (18). An important change in pesticide application during the study period may have been the change from open
cabins to closed cabins on tractors. However, the most important changes in application methods occurred after the study period, starting in the late 1990s. We used a general JEM-based
estimate of pesticide exposure (no/low/high), but the specific
intensity of exposure may depend on the prudence and (protective) equipment of the pesticide applier. Moreover, the specific
mechanism leading to damage in the lungs is probably different
for each pesticide; the specific mechanism, among others, depends on the affected biochemical pathway and on the vapor
and aerosol droplet size. For example, the primary mechanism
for paraquat toxicity is its cyclic redox reaction and consequent
free radical generation, resulting in oxidative damage to the
lung tissue (19). It is very well likely that the effect of exposure
to such a pesticide is more pronounced when antioxidant systems are already depleted by cigarette smoking and the lung tissue has already been damaged by the free radicals from tobacco
smoke. Occupational exposure to pesticides may then act synergistically with tobacco smoke exposure, as suggested by the
interaction between smoking and pesticide exposure that we
found in both the previous cross-sectional study (5) and the current longitudinal study.
Sunyer et al. (10), using the same JEM to assess occupational
exposure, did not detect an accelerated decline in FEV1 among
subjects exposed to dusts, gases, and fumes or the composite
measure VGDF in a relatively young population (ages 25–40

Pesticide Exposure and Lung Function Decline 1329

Am J Epidemiol. 2014;179(11):1323–1330

billion workers worldwide (about 34% of the global workforce) (22); therefore, health effects associated with occupational exposure to pesticides can have a large public
health impact. This is especially true in populations that are
highly exposed, such as agricultural workers in developing
countries, who often apply pesticides with insufficient protective equipment and training (23).
In conclusion, we have shown that occupational exposure
to pesticides is associated with clinically relevant accelerated
annual decline in lung function in the general population.
This may subsequently increase the risk for development of
COPD and thereby contribute to the large burden of morbidity and mortality associated with this disease.

ACKNOWLEDGMENTS

Author affiliations: Department of Epidemiology, Faculty
of Medical Sciences, University of Groningen, University
Medical Center Groningen, Groningen, the Netherlands (Kim
de Jong, H. Marike Boezen, Judith M. Vonk); Groningen
Research Institute for Asthma and COPD, University of
Groningen, University Medical Center Groningen, Groningen,
the Netherlands (Kim de Jong, H. Marike Boezen, Dirkje
S. Postma, Judith M. Vonk); Division of Environmental Epidemiology, Institute for Risk Assessment Sciences, University of
Utrecht, Utrecht, the Netherlands (Hans Kromhout, Roel
Vermeulen); and Department of Pulmonology, Faculty of
Medical Sciences, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands (Dirkje
S. Postma).
The work of Kim de Jong was funded by Research Institute
GUIDE, University of Groningen, University Medical Center
Groningen. The Vlagtwedde-Vlaardingen cohort study was
supported by the Ministry of Health and Environmental Hygiene of the Netherlands and the Netherlands Asthma Fund
(grant 187).
The sponsors of this study played no role in the design of
the study; in data collection, analysis, and interpretation; or in
the writing and submission of the manuscript.
The University of Groningen has received money for D.S.P.
regarding an unrestricted educational grant for research from
AstraZeneca (AstraZeneca plc, London, United Kingdom)
and Chiesi Pharmaceutical (Chiesi Farmaceutici S.p.A.,
Parma, Italy). Fees for consultancies were given to the University of Groningen by AstraZeneca, Boehringer Ingelheim
(Boehringer Ingelheim GmbH, Ingelheim am Rhein,
Germany), Chiesi, GlaxoSmithKline (GlaxoSmithKline plc,
London, United Kingdom), Nycomed (Takeda Pharmaceuticals International GmbH, Zurich, Switzerland), and Teva
(Teva Pharmaceutical Industries Ltd., Petach Tikva, Israel).
H.K. has been a consultant to the Norwegian government.
The other authors declare that they have no conflicts of interest.

REFERENCES
1. Balmes J, Becklake M, Blanc P, et al. American Thoracic
Society statement: occupational contribution to the burden of

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

years). Our cohort consisted of a more heterogeneous sample of
older persons who had lung function measurements taken every
3 years, yet the association between exposure to VGDF in the
last-held job and annual decline in FEV1 in our study disappeared after adjustment for pesticides. There was a marginally
significant association between cumulative exposure (intensityyears) to VGDF and annual decline in FEV1.
Our study had several strengths and limitations. Our general
population was followed for 25 years, and we had repeated
measurements for the majority of subjects included; on average,
5 measurements were available for each subject. Subjects were
included from a both rural area (Vlagtwedde) and an urban area
(Vlaardingen), yet effect estimates remained similar when we
adjusted for area (results not shown). Secondly, we used 2 estimates of occupational exposure: exposure in the last-held job
and cumulative exposure (intensity-years) based on number of
years exposed in the last-held and 2 previous jobs reported at
the last measurement, weighted by intensity of exposure. Duration of employment in the reported jobs was available for
93% of the subjects. Conclusions remained similar when we
assumed exposure of 1, 5, or 10 years when information on
job duration was missing. Moreover, using estimates of occupational exposure in the last-held job and using estimates of
cumulative exposure (intensity-years) resulted in the same conclusions. Finally, exposures were estimated with the ALOHA+
JEM, which was specifically designed for general-populationbased studies. In general, an advantage of using JEM-based
exposure estimates rather than self-reported exposures is that
they are less likely to be affected by recall bias and differential
misclassification of exposure (20, 21).
An important limitation of our study is that we were not able
to completely disentangle the associations of VGDF and pesticide exposure; that is, high pesticide exposure was always accompanied by high VGDF exposure. In the analyses based on
joint high levels of exposure to VGDF and pesticides, we confirmed that there was no association for only high VGDF exposure. It remains to be elucidated whether the association
found with pesticides is driven by an association for pesticide
exposure only or an association for joint exposure to pesticides
and VGDF. Moreover, we did not have specific information
on the pesticides applied; this information may be required
to determine the specific biological pathways by which pesticides affect decline in lung function and consequently to set up
interventions. In addition, since the late 1990s, there have been
changes in regulation of application methods and pesticides
available on the market; for example, paraquat has been
banned in the European Union since 2007. Also, fewer people
smoke today than during the study observation period. Therefore, translation of these findings to today’s workers is uncertain and will require studies in cohorts of historically younger
age. Finally, occupational exposure was ascertained at the last
study visit in 1989/1990. Subjects with respiratory complaints
or lung function impairment could have changed occupations
or dropped out of the study. This may have resulted in underestimation of the associations between occupational exposures
and annual decline in lung function.
Notwithstanding, with the current study we have shown
that occupational exposure to pesticides is associated with
clinically relevant accelerated decline in both FEV1 and
FEV1%VC. The agricultural sector employs more than 1.1

1330 de Jong et al.

2.
3.

4.
5.

7.
8.

9.
10.
11.

12. van der Lende R, Kok T, Peset R, et al. Longterm exposure to
air pollution and decline in VC and FEV1. Recent results from a
longitudinal epidemiologic study in the Netherlands. Chest.
1981;80(suppl. 1):23–26.
13. Rijcken B, Weiss ST. Longitudinal analyses of airway
responsiveness and pulmonary function decline. Am J Respir
Crit Care Med. 1996;154(6):S246–S249.
14. Hoppin JA, Valcin M, Henneberger PK, et al. Pesticide use and
chronic bronchitis among farmers in the Agricultural Health
Study. Am J Ind Med. 2007;50(12):969–979.
15. Valcin M, Henneberger PK, Kullman GJ, et al. Chronic
bronchitis among nonsmoking farm women in the Agricultural
Health Study. J Occup Environ Med. 2007;49(5):574–583.
16. Peiris-John RJ, Ruberu DK, Wickremasinghe AR, et al. Lowlevel exposure to organophosphate pesticides leads to restrictive
lung dysfunction. Respir Med. 2005;99(10):1319–1324.
17. Cha ES, Lee YK, Moon EK, et al. Paraquat application and
respiratory health effects among South Korean farmers. Occup
Environ Med. 2012;69(6):398–403.
18. Anwar WA. Biomarkers of human exposure to pesticides.
Environ Health Perspect. 1997;105(suppl 4):801–806.
19. Bus JS, Gibson JE. Paraquat: model for oxidant-initiated
toxicity. Environ Health Perspect. 1984;55:37–46.
20. ‘t Mannetje A, Kromhout H. The use of occupation and industry
classifications in general population studies. Int J Epidemiol.
2003;32(3):419–428.
21. Kromhout H, Vermeulen R. Application of job-exposure
matrices in studies of the general population: some clues to their
performance. Eur Respir Rev. 2001;11:80–90.
22. International Labour Organization. Global Employment Trends
2012. Geneva, Switzerland: International Labour Organization;
2012.
23. World Health Organization. Public Health Impact of Pesticides
Used in Agriculture. Geneva, Switzerland: World Health
Organization; 1990.
24. Quanjer PH, Tammeling GJ, Cotes JE, et al. Lung volumes and
forced ventilatory flows. Eur Respir J. 1993;6(suppl. 16):5–40.

Am J Epidemiol. 2014;179(11):1323–1330

Downloaded from http://aje.oxfordjournals.org/ at Universidade de São Paulo on August 14, 2014

6.

airway disease. Am J Respir Crit Care Med. 2003;167(5):
787–797.
Zock JP, Sunyer J, Kogevinas M, et al. Occupation, chronic
bronchitis, and lung function in young adults: an international
study. Am J Respir Crit Care Med. 2001;163(7):1572–1577.
Mehta AJ, Miedinger D, Keidel D, et al. Occupational exposure
to dusts, gases, and fumes and incidence of chronic obstructive
pulmonary disease in the Swiss Cohort Study on Air Pollution
and Lung and Heart Diseases in Adults. Am J Respir Crit Care
Med. 2012;185(12):1292–1300.
Matheson MC, Benke G, Raven J, et al. Biological dust
exposure in the workplace is a risk factor for chronic obstructive
pulmonary disease. Thorax. 2005;60(8):645–651.
de Jong K, Boezen HM, Kromhout H, et al. Pesticides and other
occupational exposures are associated with airway obstruction:
the LifeLines cohort study. Occup Environ Med. 2014;71(2):
88–96.
Blanc PD, Iribarren C, Trupin L, et al. Occupational exposures
and the risk of COPD: dusty trades revisited. Thorax. 2009;
64(1):6–12.
Iversen M, Dahl R. Working in swine-confinement buildings
causes an accelerated decline in FEV1: a 7-yr follow-up of
Danish farmers. Eur Respir J. 2000;16(3):404–408.
Bakke B, Ulvestad B, Stewart P, et al. Cumulative exposure to
dust and gases as determinants of lung function decline in
tunnel construction workers. Occup Environ Med. 2004;61(3):
262–269.
Wang X, Zhang HX, Sun BX, et al. Cross-shift airway
responses and long-term decline in FEV1 in cotton textile
workers. Am J Respir Crit Care Med. 2008;177(3):316–320.
Sunyer J, Zock JP, Kromhout H, et al. Lung function decline,
chronic bronchitis, and occupational exposures in young adults.
Am J Respir Crit Care Med. 2005;172(9):1139–1145.
van Diemen CC, Postma DS, Vonk JM, et al. A disintegrin and
metalloprotease 33 polymorphisms and lung function decline in
the general population. Am J Respir Crit Care Med. 2005;
172(3):329–333.


Related documents


am j epidemiol 2014 de jong 1323 30
rauh et al 2011
eskenazi et al 2004
3 jensen final
news release ucla combined pesticides
6 brading et al 2009


Related keywords